Python is alive! With such a loud statement I want to start an article.

This week came the 100th Python Digest. On this occasion, we decided to summarize the work on the digest and talk about the trends identified in the process of collecting news on the language, which is actually dedicated to the digest.


')
At the end of 2014, we summed up the year. This time we will look at trends in less than 11 months of 2015.

But before moving on to trends, let us tell you what we have done in (almost) a year.

Those interested in the details of mercy, we ask under the cat.

This year was difficult for us - in the summer the Digest could cease to exist, because the founder of the project plunged into the abyss of work and physically could no longer moderate the news.
However, the new wave of forces in the person of Alexander Sapronov (@WarmongeR) forced the project to live and develop further.

Taking the initiative, I began to actively automate the moderation of news. Added a lot of new features, dragged a digest to Python 3, corrected quite obvious project errors. Expanded content, adding not a small number of new sources. The digest began to aggregate video releases libraries.

Now we have a stable attendance, and even in twitter there are more than 550 subscribers, we began to communicate with readers in gitter, and the soulless bot publishes news in the Telegram channel. However, the end of the year was also saddened by the digest - after 96 issues, due to the workload in our own projects, we were no longer able to draw pictures for the releases from Andrey (@ owlman75). Is anyone ready to pick up this honorary flag? If yes - write in a personal or in the comments.

On that note, let's get to the news. And we will tell you about ...

Python and its PEPs

The September release of Python 3.5 is the main news in the Python universe. However, there were other interesting events - Guido told about the first years of Python , Django was 10 years old , Django Girls was one year old , pip was updated to version 7.0 and learned how to cache modules. But what to hide - PyCharm 5.0 was released .

The digest flashed other interesting links:

• Tutorials: What's new in Python 3.5 - When are type annotations useful? Will asynchronous programming become common practice with new async-await? Arrange a discussion of these and other new products Python 3.5. All this you can learn in the report of Andrei Vlasov
• PEP 0506 - Adding the secrets module to the standard library - The secrets module contains several pseudo-random number generators. In 3.6, this module will be added to the standard library.
• PEP-498 - a convenient way to format strings - This PEP greatly enhances the ability to form strings. Study this standard, it will break your brain.
• Python 3.5 and multitasking - In Python 3.5, asyncio has become more native. In the article you will find a comparison of the speed of work with different configurations of iron.
• Counting the elements in the list: how it was implemented before, and how now - a guided history of Python

Releases and projects

Over the past year, several major updates happened - IPython was divided and updated to version 3.0 , release python-oauth2 after 4 years of inactivity (added Python 3). Python 3 support added to gevent .
There were more interesting releases and projects:

• Past, Present and Future of Jupyter and IPython - Recently IPython was divided into two parts - Jupyter and IPython (to increase modularity). In this video you will find the latest ideas on this topic.
• Pandas, starting with version 0.17, releases GIL - Like Numpy, Pandas now knows how to release / raise GIL. This allows you to increase productivity.
• SQLAlchemy 1.0.0 - The result of a ten-year, active development of a module for abstracting data from databases and ORM. There was, however, no immediate release 1.0.1
• Search for duplicate pictures Python - The article describes an algorithm for finding duplicates among pictures. There you can find a working code.
• hask - Haskell in Python - If you like functional programming, then you should pay attention to this library. It implements the many features of the Haskell language.
• MesoPy - getting weather data from over 40,000 stations - MesoPy is a wrapper over the MesoWest API, which provides the ability to collect weather data from many weather stations.

Meetings and Conferences

For the whole year, Python-developers visited and organized many different conferences and meetings - the Minsk Pythonists all also delight us with wonderful meetings, the new meeting season MoscowPython, the organization of the second Python-nist community in St. Petersburg (SPbPy and SPb Python) and the first in Novosibirsk (PyNSK) - all this and much more pleased us:

• The best reports of PyCon 2015 - the link leads to a set of videos that are worth watching
• PiterPy conference report # 2 - March 20, 2015 The 2nd annual PiterPy conference was held. The author of the post offers a review of this event with all the necessary links.
• Europython 2015 Conference Report
• Overview of masterclasses and tutorials from PyCon Montreal 2015 - Programs for such large conferences as PyCon are too extensive to quickly navigate. Articles similar to this one help a lot.
• Video Conferences Dnepr Py # 4 - Report from the Conference DneprPy

Habrapiton

A significant contribution to the Russian-language Python-content makes Habr.
Since the beginning of the year, we have followed interesting articles and even article cycles - Event-oriented Python backtesting step by step , Log Analysis with Hadoop / Python , Django 1.9 will receive a new admin design , Cost of housing as a function of coordinates , Immersion into the Python interpreter. P1 .

I would like to mention other interesting articles from Habr:

• Celery: best practices
• Remote execution of system commands on request via sockets in Python 3. Data transfer protocol
• Redirecting data from the COM port to the Web - Recently, an article titled “Displaying data from Serial in Chrome Application” was on Habré about how beautifully to present data sent by Arduin to Serial. In the comments to the article, it was regretted that such a solution would not work under Firefox and the idea was expressed that “you can still write a simple web server with html output based on this thing”.
• Automatic migrations to Peewee
• Dictionary implementation in Python 2.7
• Implementing a search engine with Python rankings (Part 2) - In the previous section, we built an index, but we still cannot execute queries on it. About this I will tell in this article.

Analyze it!

Data analysis and visualization is the biggest trend in Python over the past year. All summer, the authors compared Python and R - what kind of performance, what syntax, what libraries are there, and even compared the number of lines to implement the same analysis.

The most important events of this year in this section were the division of IPython into Jupyter and the notebook and Pandas themselves learned to lower the GIL - which makes it easier to work with multithreading.

But there were other news - analyzed a variety of data:

• Analyzing 1.7 billion comments with Reddit using Blaze and Impala - An exciting article about data analysis. The peculiarity of this article is the volume of data. Far from any car such a volume will go down in memory.
• Visualization of the movements of NBA players in Python - An interesting article that shows how to visualize the movement of NBA players, as well as how to calculate various statistics on this data.
• Data Analysis: Crime in Baltimore - The article shows the application of data analysis on the example of the crime rate in the city of Baltimore. The tools are standard - pandas, matplotlib, ipython, but the data is not standard.
• Draw a map of British pubs

And they made a lot of educational materials:

• Introduction to Statistics - A rather large book on building statistical reports using Python got to version 7.1 this week. The author constantly keeps the text and code of the book up to date.
• Collection of tutorials for data processing - The article provides a list of the 30 best Data Scientists, as well as an excellent list of guides on various aspects of data processing.
• intro2stats - introduction to statistics in Python

Machine learning

People are lazy creatures, and developers are even lazier. They make the soulless machine do everything for them. Including through machine learning:

• Introduction to machine learning in Python - In this guide you will find the basic basics of machine learning, short examples, and terms.
• Simple genetic algorithm in 15 lines
• Neural network in 13 lines. Part 2
• neupy - a library for Neural Networks
• word2vec & doc2vec: Python Key Analysis
• Selection of recommendations based on the analysis of links on Twitter using Neo4j - A rather interesting, detailed article with illustrations and examples of real code.
• We imitate the writing style with the help of Markov chains - The article describes how you can use Markov chains to write a text in a special style.
• Polyglot NLP - a module for processing texts in many languages - This module supports the basic word processing algorithms. Chip module in support of a huge number of languages.
• Real-time motion analysis using machine learning

Just wondering

As fate would have it, the Python developers fell in love with the functional approach and from time to time try to shift the development of the language in this direction.

• Functional Python Programming - introductory article on functional Python programming
• effect - functional library, adds isolation of side effects
• hask is a functional Python programming library that tries to repeat Haskell

But not only the functionalism visited the head of the pythonists. As always, they were engaged in profiling, micro-optimization, studied data structures:

• Is my python code slow? Profiling tips. - The article displays the minimum set of actions that are needed to find slow spots in the Python code.
• Best of the best practices for Python
• Micro code optimization - we speed up the search for variables in scope

But what amused us is the change of facial features on the fly:

• Changing facial features using Python - The article presents a script that automatically allows you to take the facial features of one person and set them to another person. Dlib and OpenCV are used for this.
• Automatic face changer ... now on video

Iron Python

This year was weak on iron in Python - Raspberry Pi and other unicameral computers have been studied from head to toe, and Intel Edison is too expensive for home crafts. However, among bezrybya found an interesting use of crafts:

• Using Python with Arduino - 17 lessons with text and video with a fairly deep immersion in the topic of software development for the popular board constructor
• Manage TV with voice and Raspberry Pi - Have you ever dreamed of implementing automatic channel switching when advertising starts? Then this article will help you.
• How I hacked the WiFi button for $ 5 to track the child - The author of the article tells how he uses the $ 5 WiFi button from Amazon to track the child - he collects statistics on how much the child slept

This year, developers are fond of computer vision on OpenCV:

• Did I turn off the oven? - We treat the evil eye according to the photo. We determine the status of the oven from the photo - on / off.
• We track the ball on OpenCV - Recognizing an object on a video is not an easy task. This article shows how to search for the ball on video.
• We play Pacman with gestures: Python + OpenCV . In the article you will find instructions on how to implement gesture processing in Python. As an example, the game was chosen Pacman, in which the author and played with gestures.

What's next

I would like to reflect on the future of the project Python Digest.

First, the project will live on. Despite the lack of time among the team members, we are developing the project and still popularize Python in the Russian language segment of the Internet.

Secondly, we are looking for new strengths among readers. Open to suggestions and changes. We happily expect a new illustrator, developers, news moderators, authors.

Thirdly, we are looking for financial support from the community and companies, for this reason the button “Donate” appeared on the site.